Oxygen treatment for cluster headache attacks at different flow rates: a double-blind, randomized, crossover study

BackgroundCluster headache attacks can, in many patients, be successfully treated with oxygen via a non-rebreather mask. In previous studies oxygen at flow rates of both 7 L/min and 12 L/min was shown to be effective. The aim of this study was to compare the effect of 100% oxygen at different flow rates for the treatment of cluster headache attacks.MethodsIn a double-blind, randomized, crossover study, oxygen naive cluster headache patients, treated attacks with oxygen at 7 and 12L/min. The primary outcome measure was the percentage of attacks after which patients (treating at least 2 attacks/day) were painfree after 15min, in the first two days of the study. Secondary outcome measures were percentage of successfully treated attacks, percentage of attacks after which patients were painfree, drop in VAS score and patient preference in all treatment periods (14days).ResultsNinety-eight patients were enrolled, 70 provided valid data, 56 used both flow rates. These 56 patients recorded 604 attacks, eligible for the primary analysis. An exploratory analysis was conducted using all eligible attacks of 70 patients who provided valid data. We could only include 5 patients, treating 27 attacks on the first two days of the study, for our primary outcome, which did not show a significant difference (p=0.180). Patients tended to prefer 12L/min (p=0.005). Contradicting this result, more patients were painfree using 7L/min (p=0.039). There were no differences in side effects or in our other secondary outcome measures. The exploratory analysis showed an odds ratio of being painfree using 12L/min of 0.73 (95% CI 0.52-1.02) compared to 7L/min (p=0.061) as scored on a 5-point scale. The average drop in score on this 5-point scale, however, was equal between groups. Also slightly more patients noticed, no or not much, relief on 7L/min, and found 12L/min to be effective in all their attacks.ConclusionThere is lack of evidence to support differences in the effect of oxygen at a flow rate of 12L/min compared to 7L/min. More patients were painfree using 7L/min, but our other outcome measures did not confirm a difference in effect between flow rates. As most patients prefer 12L/min and treatments were equally safe, this could be used in all patients. It might be more cost-effective, however, to start with 7L/min and, if ineffective, to switch to 12L/min.